Knocking on the Outgroup's Door: Seeking Outgroup Help Under Conditions of Task or Relational Conflict

Abstract

Three studies investigated the willingness to seek help from another group in situations where collaborative goals are undermined by task or relational conflicts between the groups. Compared to task conflict, relational conflict was argued to trigger a striving for more autonomy. The results from three experiments (N = 82, N = 65, and N = 62) supported the prediction that relational conflict, compared to task conflict, promotes more help avoidance, in particular avoidance of dependency-oriented help (a full solution). As expected, no difference was found for the willingness to seek autonomy-oriented help (a hint) from the other group.

ACKNOWLEDGMENTS

Part of this research was facilitated by an innovation grant VENI from the Netherlands Organization for Scientific Research awarded to the first author.

Notes

¹Literature on intragroup conflict also employs the term task conflict, which refers to disagreement about the distribution of resources, procedures, or outcomes related to the task (e.g., De Dreu & Weingart, 2003). In the current article, task conflict is defined in line with literature on intergroup conflict and refers to incompatible group goals.

²Following the figure estimation task, a brief word completion task was administered after which half of the participants received positive group feedback (60% success) and half received negative group feedback (41% success), the purpose of which was to explore group status differences in help seeking. However, checks showed that this manipulation was unsuccessful, and it did not affect any of the dependent variables reported here, neither as a main effect nor in interaction with the conflict manipulation (all Fs < 1). A possible explanation is that the word completion task was of a different nature than the subsequent knowledge task, which was used to assess help seeking behavior. If participants perceived the two tasks as unrelated, a status difference on one dimension is unlikely to transfer to the other dimension. In addition, the difference in feedback scores may not have been big enough to have a meaningful impact. Status was therefore not included as a factor in the analyses reported in this article.

³The knowledge questions were pretested in a comparable student sample. On the basis of this pretest, 10 difficult questions (where most people indicated that they did not know the answer), 10 moderately difficult questions (where most people indicated that they believed they knew the answer but weren't sure about it), and 10 easy questions (where most people indicated that they knew the answer and were sure about it) were selected. Questions in which solution probability was influenced by sex or study major were excluded.

⁴Because it is known that people adjust their help-seeking behavior to their beliefs about the likelihood that a request for help will be granted (Wills & DePaulo, 1991), requests for help in this study could not be rejected by the outgroup. This might lead to an overestimation of the amount of requested help in general but will have no bearing on differences between conditions, which is the primary interest in this study. Moreover, this situation parallels many societal settings in which requests for help are guaranteed to be met, for example, when low-income people are entitled to government subsidies provided that they put in a request, which usually comes at no additional costs.